The response model describes the response to fetching a prediction.
You wyoul need to build a required using the request model.
Thats in broadcaster_id and the token used to make the request.
But this is your best bet for help with a specific library.